This is typically the topic of most concern when initially considering an employment offer. As you begin to seek new employment opportunities, you may have a pretty good idea what salary to expect (based on discussions with colleagues, published salary surveys, or other sources) or you may enter the process completely open-minded, ready to compare multiple offers as they are presented to you.
This topic is made more complicated because compensation models are evolving and may differ during the term of your contract. It is not uncommon for compensation to start out as a fixed salary (generally paid on a biweekly basis), then transition into one partially or wholly based on productivity and other performance measures.
The employer has a need for your expertise or wouldn’t be expending the resources necessary to recruit you. Employers are generally prepared to offer a fair salary based on salary surveys, current local salary ranges and other insights that their human resources departments provide. But they also have to consider other factors. When I was recruiting for our hospital-based multi-specialty practice, I had many concerns:
- When, if ever, will collections (and even down stream revenues) justify the salary, benefits and other expenses associated with this hire?
- Will the new recruit be a cultural fit with the organization?
- Is the new physician truly interested in entering into a long-term relationship, or will we be investing in her only to see her leave in a year or two?
- Does the new physician understand what it takes to build a new practice and is she prepared to assist in that process?
- Will the new recruit work well with other team members, including staff, managers, leaders and other medical staff?
- Does the physician understand the importance of good documentation and is he committed to balancing quality, access and cost?
You should keep these concerns in mind as you negotiate the terms of your employment.
Understanding Salary Surveys
A salary survey is just that: a survey. It is a inquiry sent by the surveying entity to obtain salary information from a sampling of physicians, either directly (i.e., from specialty society members), or from employers. They are therefor subject to certain biases. I’ve not read any studies that quantify the biases, but it seems that some are obvious. For example, in general, hospital systems tend to offer higher salaries to new graduates than physician groups for many specialties. And some of the salary surveys, such as the Medical Group Management Association (MGMA) survey, have a higher representation of responses from such hospital based physician groups, while others (such as the American Group Management Association) are more representative of physician owned groups. It’s less likely that a physician owned medical group will make an offer equal to the median salary suggested by the MGMA survey.
The survey might have a small number of responses in certain specialties and geographic areas, so anchoring your expectations to that number might not be justified.
The surveys generally present both salary means and salary medians. Employers and recruits tend to focus on the medians because they are less impacted by outliers. The surveys, in addition to listing means and median salary levels, generally provide salary data at the 25th, 75th, 90th and 95th percentiles. Other information that may be included in such surveys is gross revenues, collections, office expenses, and worked RVUs.
Understanding Worked RVUs
During and after the contracting process, worked relative value units (wRVUs) are sure to be discussed, and need to be understood. wRVUs are used to describe and measure the productivity of physicians. They are often used to determine the size of a bonus payment or even 100% of a physician’s salary.
Before RVUs were developed and applied to physician payments, Medicare paid for services based on usual and customary fees, which were highly variable and arbitrary. In order to standardize payments, the Health Care Financing Administration (now known as the Centers for Medicare and Medicaid Services, or CMS) adopted the resource based relative value scale (RBRVS) system after it was development in the late 1980s. A team led by William Hsiao, Ph.D., at Harvard University, published its RBRVS in the Journal of the American Medical Association (JAMA) in 1988. Its use was formally signed into law in 1989 and began to be used in 1992.
The RBRVS, as adopted by CMS, is used to pay for medical services, including medical care provided in hospitals, clinics, emergency rooms, nursing homes and elsewhere. The RBRVS is comprised of three parts: physician work, practice expense and professional liability costs. Geographic variability can be accounted for in the latter two of these factors. Medicare payments are determined by applying a dollar multiplier to the relative value of a procedure (generally between 0.0 and 30.0).
The physician work part of the formula (wRVU) is what is used to compare the productivity of physicians and to create compensation models (NOT the total or tRVU). The wRVU results from consideration of its components: the time needed to deliver a service, the relative mental effort and judgment required, and the intensity as it relates to the risk to the patient. The complete list of tRVUs and their components (including the wRVUs) are updated and published annually by CMS (Physician Fee Schedule – January 2015 Release), which also adjusts the multiplier in order to create the payment schedule for physician payments each year. To get an idea of the relative weight of various types of services, I am including a very small sample few of wRVUs and their weights below:
|Sample of wRVUs from the 2015 Medicare RBRVS|
|HCPCS (E/M Code)||DESCRIPTION||WORK RVU|
|92979||Intravasc us heart add-on||1.44|
|92986||Revision of aortic valve||22.85|
|92987||Revision of mitral valve||23.63|
|92990||Revision of pulmonary valve||18.27|
|99201||Office/outpatient visit new||0.48|
|99202||Office/outpatient visit new||0.93|
|99203||Office/outpatient visit new||1.42|
|99204||Office/outpatient visit new||2.43|
|99205||Office/outpatient visit new||3.17|
|99211||Office/outpatient visit est||0.18|
|99212||Office/outpatient visit est||0.48|
|99213||Office/outpatient visit est||0.97|
|99214||Office/outpatient visit est||1.50|
|99215||Office/outpatient visit est||2.11|
|99217||Observation care discharge||1.28|
Notice that a typical new patient office visit of moderate complexity (99204) runs about 2.43 wRVUs. An established patient office visit at the same level runs only 1.5 wRVUs. Operative procedures run much higher values. Using these weighted values, it is possible to compare the productivity of physicians in the same specialty, and even those in different specialties. Typically, an internist or family physician will generate 4,000 to 5,000 wRVUs per year, or roughly 100 wRVUs per week (assuming 48 weeks of work). If the average visit has weighting of 0.97 wRVUs, then there will be about 97 visits. If the average weighting is closer to 2.0, then the number of visits will be closer to 50 per week, but they will be longer, more complex visits.
A general surgeon may easily generate 7,000 to 8,000 wRVUs; an invasive cardiologist or neurosurgeon 10,000 or more, primarily because of the heavy weighting of procedural visits/codes.
While initial salaries may not be linked directly to wRVUs, there will usually be an expectation by the employer that the RVUs being generated (through patient visits and procedures) will begin to meet or exceed the wRVUs of similarly compensated physicians in the national or regional survey within a year or two of being hired. For example, consider a hypothetical Dr. Smith who was hired at a salary equal to the median salary for his specialty of $160,000 per year. Two years later he seeing about 50 patients per week and generating 3,500 wRVUs. Unfortunately, this RVU level correlates with the 25th percentile level for his specialty in the survey. He will probably be generating revenue and net collections for his employer at the 25th percentile as well, which probably does not cover the costs of his salary, benefits and overhead expenses. So, he seems to be overpaid and this will need to be reconciled in some manner.
There are many other similar nuances to the use of RVUs that can render their use in employment contracts somewhat challenging. wRVUs may be used to define bonus thresholds. But what if the work that is being done does not generate visit codes for the physician, and therefore would not be reflected in a bonus calculation.
For example, consider a situation in which an internist will be paid a bonus in any given year for wRVUs personally generated and billed above a threshold of 4500. In addition to seeing patients in the office and the hospital, she is collaborating with an NP that sees overflow patients and she is working a few hours a week as a medical director for a nursing home as part of her work duties. Since these additional duties do not generate billable visits and RVUs for the physician, she will not be compensated for those duties under an RVU bonus model, even though those duties take her time and effort to accomplish.
I am inserting the links below for those that want to read more about the use of RVUs in employment contracts. I will also be devoting future posts to specific examples of the unforeseen consequences of the use of wRVUs to structure compensation for physicians.